The Research Of The Automated Face Detection Methods Based On The Statistical Theory

With the development of the computer and information technology, image processing gets much more widely used in many fields, such as science, national defense, industry and commerce, finance, et. Face detection and recognition research is related to the fields of pattern recognition, image processing, physiology, cognitive science and so on. Among the features measured, facial feature identification and verification are gaining popularity and diverse applications for the reason that they are considered to be non-invasive, low cost, and natural biometric technologies. At the same time, it is tied with other researches such as Biometric Authentication and Human-Computer Interaction. It has become one of the key issues because of its practical essentiality in many aspects, especially in the field of security. The main contents of the thesis are listed as followings:1.The theory of Support Vector Machines was proposed under the Statistical Learning theory which based on Structural Risk Minimization Principle. The method of SVM gets better generalization than Artificial Neural Networks(ANN), which based on Empirical Risk Minimization Principle. And it was proved to be applicable in complicated face detection problems. Support Vector Machine is made of training section and detecting section. A lot of face samples and " not face" samples are used to train the SVM classifier, to get optimal separating hyperplane in the training. And SVM classifier is used to detect faces in the detecting. This algorithm can be used to detect many frontal faces with different sizes in a color image or gray image.2. Concepts of rectangle features and"integral image"are introduced. We use Adaboost to build a robust scale-invariant image classifier for face detection. And based on the basic principle of Adaboost algorithm and rectangle features, the cascade classfiers for face detection is constructed. Finally, the performance of the classifers is tested.3. We introduce a method of face recognition. Kernel Principal Component Analysis (KPCA) is a non-lineal generalization of PCA, which extracts non-linear information from human facial image.The experiment results show that these methods have good detection rates.